Upload 2 files

app.py

@@ -0,0 +1,90 @@
+import streamlit as st
+import speech_recognition as sr
+from deep_translator import GoogleTranslator
+from pydub import AudioSegment
+from io import BytesIO
+import tempfile
+
+# Title of the app
+st.title("Speech-to-Text with Translation to English")
+
+# Initialize recognizer
+recognizer = sr.Recognizer()
+
+# Choice for input method
+input_method = st.radio("Select Input Method", ("Record from Microphone", "Upload Audio File"))
+
+# Choice for input language
+language_options = {"English": "en", "Hindi": "hi"}
+input_language = st.selectbox("Select Input Language", options=language_options.keys())
+selected_lang_code = language_options[input_language]
+
+# Function to convert audio chunk to text
+def speech_to_text(audio_data, lang="en"):  # Default language to English
+    try:
+        # Recognize speech
+        st.info("Converting speech to text...")
+        detected_text = recognizer.recognize_google(audio_data, language=lang)
+        return detected_text
+    except Exception as e:
+        st.error(f"Error in speech recognition: {e}")
+        return None
+
+# Handle recording from microphone
+if input_method == "Record from Microphone":
+    if st.button("Start Recording"):
+        with st.spinner("Recording... Please speak into the microphone."):
+            try:
+                # Capture audio input from the microphone
+                with sr.Microphone() as source:
+                    st.info("Listening... Please speak now.")
+                    recognizer.adjust_for_ambient_noise(source)  # Adjust for background noise
+                    audio_data = recognizer.listen(source)
+                    st.success("Recording complete!")
+                
+                # Process and convert speech to text
+                detected_text = speech_to_text(audio_data, lang=selected_lang_code)
+                if detected_text:
+                    st.write("Detected Speech Text:", detected_text)
+                    
+                    # Translate to English
+                    translator = GoogleTranslator(source='auto', target='en')
+                    translated_text = translator.translate(detected_text)
+                    st.write("Translated Text (English):", translated_text)
+                    
+            except sr.UnknownValueError:
+                st.error("Could not understand the audio. Please try again.")
+            except sr.RequestError as e:
+                st.error(f"Could not request results from the service; {e}")
+
+# Process uploaded audio file
+if input_method == "Upload Audio File":
+    uploaded_file = st.file_uploader("Upload an audio file", type=["wav", "mp3", "ogg"])
+    if uploaded_file:
+        with st.spinner("Processing uploaded audio..."):
+            try:
+                # Convert uploaded file to WAV format using pydub
+                audio = AudioSegment.from_file(BytesIO(uploaded_file.read()))
+                # Split audio into 30-second chunks
+                chunk_duration_ms = 30000
+                chunks = [audio[i:i+chunk_duration_ms] for i in range(0, len(audio), chunk_duration_ms)]
+                text_output = ""
+
+                for i, chunk in enumerate(chunks):
+                    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_wav_file:
+                        chunk.export(tmp_wav_file.name, format="wav")
+                        with sr.AudioFile(tmp_wav_file.name) as source:
+                            audio_data = recognizer.record(source)
+                            detected_text = speech_to_text(audio_data, lang=selected_lang_code)
+                            if detected_text:
+                                text_output += detected_text + " "
+
+                # Display detected text and translate
+                if text_output:
+                    st.write("Detected Speech Text:", text_output)
+                    translator = GoogleTranslator(source='auto', target='en')
+                    translated_text = translator.translate(text_output)
+                    st.write("Translated Text (English):", translated_text)
+
+            except Exception as e:
+                st.error(f"Error processing the audio file: {e}")

requirements.txt

@@ -0,0 +1,4 @@
+streamlit
+speechrecognition
+deep-translator
+pydub